Nathan F
sage
Reged: 10/10/08
Posts: 223
|
|
Is it possible to build a truss dob that has absolutely zero perceptible collimation drift - as measured with an autocollimator - from horizon to zenith? Are there certain mirror cell designs, truss/strut configurations, etc.. that you have found contributes to or obtains this goal?
I'd love to hear everyone's solution to this issue.
--------------------
12 inch f/5 Deep Space Observer Dob
31mm Nagler, 13mm Ethos, 1.6x Antares 2 inch Barlow
Celestron SkyMaster 15x70 Binoculars
Homemade Maple Parallelogram Binocular Mount
|
RossSackett
professor emeritus
Reged: 08/17/07
Posts: 692
Loc: Memphis, TN
|
|
Absolutely none? Of course not--the physics won't let you.
Not enough to matter? The brute force way would be to beef up the structure and go to exotic lightweight materials like carbon fiber, but the expense and inconvenience would probably make you wish you hadn't. And every gram of structure you add also adds to the force acting to distort it, so there's an escalating arms race going on.
The clever solution would be something like Serrurier's design for the Palomar 200" double truss, with the truss sag balanced so that the upper tube and lower tube sagged proportionately, keeping the primary and secondary optics in alignment. You would have to actually engineer the sags, however--simply building a double truss isn't enough.
Ross
--------------------
"A craftsman relies on science when the state of knowledge allows it, tradition and experience when it does not, and makes art whenever he can."
12 scopes from 4.25 to 18" and a 24" in progress. 12 ATM awards. Webpage: http://stardazed.com/ Some more scope pix at http://www.flickr.com/photos/8315630@N04/
Anagrams: Amateur astronomer = A mature moon-starer; Dobsonian maker = Debonair as monk
|
Dick Jacobson
super member
Reged: 12/22/06
Posts: 132
Loc: Plymouth, Minnesota, USA
|
|
I built a double-truss 14" Newtonian on an equatorial mount. Both the mirror end and eyepiece end are supported by 6-pole trusses from a central ring bearing. The lower truss uses 3/4" aluminum poles, the upper truss 7/8" poles. I was hoping that the sag at both ends would somewhat cancel out, as Ross mentioned with the Serrurier design. I did not do any engineering calculations, and as it turned out the front end sags more than the back end; however the loss of alignment is acceptable to a non-perfectionist like myself. With careful design, I think it would be possible to build a lightweight scope that keeps perfect alignment.
--------------------
Equatorial Newtonians, 20" and 14"
"The purpose of life is to be defeated by ever greater things." - Rielke
|
JSeay86
super member
Reged: 04/26/09
Posts: 133
Loc: Norman, OK
|
|
What size dob are you thinking about building? It seems like the larger the dob the more carefully you will have to consider this because of the added weight.
Dick, do you have any pics of your 14"? I too am thinking about trying to build a dob with some materials I have laying around. Thanks!
--------------------
Jared
Orion XX12 Dob
Ten Acre Observatory
|
Dick Jacobson
super member
Reged: 12/22/06
Posts: 132
Loc: Plymouth, Minnesota, USA
|
|
Some details have changed since this picture was taken. The biggest change was that I built a frame to reinforce the "lazy susan" bearing at the center of the telescope. I found that the bearing was too flexible resulting in an unacceptable level of vibration. With the reinforced bearing, the telescope works very well.
--------------------
Equatorial Newtonians, 20" and 14"
"The purpose of life is to be defeated by ever greater things." - Rielke
|
Starman1
Vendor - Scope City
Reged: 06/24/03
Posts: 12230
Loc: Los Angeles
|
|
Quote:
Is it possible to build a truss dob that has absolutely zero perceptible collimation drift - as measured with an autocollimator - from horizon to zenith? Are there certain mirror cell designs, truss/strut configurations, etc.. that you have found contributes to or obtains this goal?
I'd love to hear everyone's solution to this issue.
I've come close. I admit that I haven't achieved absolutely NO drift between horizon and zenith. I have achieved NO drift between 30 degrees altitude and the zenith.
I learned a lot along the way:
A few things:
1) the spider vanes have to be piano-wire tight. You may have to reinforce the structure to allow this degree of tightness.
2) the pole pairs should not be parallel to one another. Either they should widen from bottom to top or the reverse, but parallel results in less rigidity and more sag.
3) the pole length should be adjusted to put the focuser near its in-most position when everything is in focus. This will significantly reduce sag in the focuser drawtube.
4) Poles are more rigid with a tiny amount of torsional twist in them. Completely relaxed tubes sag more.
5) The mirror cell should be over-built and heavy and the bottom support should be either a solid aluminum sling or the Mother of All Wires.
6) The mirror cell should support the mirror so there is no sag, no movement, and no flexure. This requires good design and/or heavy weight.
7) Add a tiny amount of weight by using larger poles. I calculate what it needs, then add 1/8-1/4" to the diameter. Too stiff doesn't exist.
When I first started out, I could see movement in the cheshire and sight tube as I moved the scope. Now I barely see any movement in the XLK autocollimator, and only below a 10 degree altitude.
It's probably an article I should write, but the point is, it can be done.
--------------------
Don Pensack
12.5" Truss Dob, 5" Maksutov, Fujinon Binos
Sustaining Lifetime IDA member
|
skyler
Carpal Tunnel
Reged: 08/16/06
Posts: 1672
Loc: TGPNW
|
|
Hello Don -
I was very close to building this type of design that could be mounted on a EQ mount for AP work. I abandoned the idea since I wanted something quickly and went with an Orion UK 12" F4 setup that is nearing completion. I still plan work on such a project but am learning a lot about rigidity and over-building as you suggest since mine will need to be used for AP as mentioned.
Do you use yours for AP as well and how does it perform for that purpose?
Thanks for all those pieces of personal experience wisdom; the most valuable kind.
Also, thanks to you, Dick, for posting that photo of your recent build. That is a nice piece of work and I can learn from what you have done.
S
|
Starman1
Vendor - Scope City
Reged: 06/24/03
Posts: 12230
Loc: Los Angeles
|
|
Skyler,
No AP for me, so my scope doesn't rotate--it merely moves up and down.
Achieving collimation retention when the tube also rotates (as on an EQ mount) is more difficult. You'll also have problems with mirror cell lateral movement, secondary center bolt shift, tube sag at focuser attachment, and flexure at tube holding points.
Identifying where the most change occurs can be difficult. In my experience, the secondary is most to blame, but the second biggest issue is mirror cell rigidity. I'm fortunate to work near Parks Optical, and seeing their solutions for fundamental problems has been eye-opening.
--------------------
Don Pensack
12.5" Truss Dob, 5" Maksutov, Fujinon Binos
Sustaining Lifetime IDA member
|
JSeay86
super member
Reged: 04/26/09
Posts: 133
Loc: Norman, OK
|
|
Dick, that's a great looking scope!
--------------------
Jared
Orion XX12 Dob
Ten Acre Observatory
|
sixela
Postmaster
Reged: 12/23/04
Posts: 10874
Loc: Boechout, Belgium
|
|
Quote:
2) the pole pairs should not be parallel to one another. Either they should widen from bottom to top or the reverse, but parallel results in less rigidity and more sag.
I seem to remember a long thread where we hashed this out. I can't seem to remember that as the conclusion.
Quote:
4) Poles are more rigid with a tiny amount of torsional twist in them. Completely relaxed tubes sag more.
References?
--------------------
400mm f/4.46 self made Dobsonian on Tom Osypowski equatorial platform
Orion Starblast (114mm f/4 reflector, Alt/Az)
|
Starman1
Vendor - Scope City
Reged: 06/24/03
Posts: 12230
Loc: Los Angeles
|
|
Quote:
Quote:
2) the pole pairs should not be parallel to one another. Either they should widen from bottom to top or the reverse, but parallel results in less rigidity and more sag.
I seem to remember a long thread where we hashed this out. I can't seem to remember that as the conclusion.
Actually, that was the conclusion and one of the main reasons the poles in large observatory instruments are non-parallel. 8 poles, in v-shaped arrangements, with top/bottom and side-to-side arrangements resist UTA sag primarily by the bending strength of the 2 side Vs, with the top poles on the sides carrying all the load. When the poles are non-parallel, the top and bottom Vs carry some of the load. The greater the difference in spacing between the poles at top and bottom, the greater the rigidity of the pole structure. A small angle change makes very little difference, percentage-wise, but it does improve rigidity a little bit. In the big scopes, the angles are often more greatly exaggerated, but then the weights born by the tubes are as well. It doesn't matter if the top is narrower than the bottom or vice-versa. It's the difference that counts.
Quote:
Quote:
4) Poles are more rigid with a tiny amount of torsional twist in them. Completely relaxed tubes sag more.
References?
I remember this from my auto-racing days in which torsion bars under stress were more rigid than torsion bars with no load. It seems that torsional flexibility decreases with load until the strength of the material begins to be exceeded, at which time a catastrophic failure of the material results. In the case of pole attachments, if the poles are stress-free when attached tightly, they will be less resistant to further torsional twisting that if there was a slight torsional twist imparted. That tension should result in greater resistance to bending, which, in long poles, we'd see as sag.
--------------------
Don Pensack
12.5" Truss Dob, 5" Maksutov, Fujinon Binos
Sustaining Lifetime IDA member
|
RossSackett
professor emeritus
Reged: 08/17/07
Posts: 692
Loc: Memphis, TN
|
|
In properly designed triangulated truss arrangements the poles are loaded in tension and compression, not bending. This makes the truss very stiff. I can't see how preloading the truss poles in torsion has much effect on overall stiffness, and runs the risk of curving the poles which would invite buckling and actually lower the stiffness.
Ross
--------------------
"A craftsman relies on science when the state of knowledge allows it, tradition and experience when it does not, and makes art whenever he can."
12 scopes from 4.25 to 18" and a 24" in progress. 12 ATM awards. Webpage: http://stardazed.com/ Some more scope pix at http://www.flickr.com/photos/8315630@N04/
Anagrams: Amateur astronomer = A mature moon-starer; Dobsonian maker = Debonair as monk
|
sixela
Postmaster
Reged: 12/23/04
Posts: 10874
Loc: Boechout, Belgium
|
|
Quote:
Actually, that was the conclusion and one of the main reasons the poles in large observatory instruments are non-parallel.
Not the way I read it. The conclusion is that it usually increased the stiffness only marginally (roughly 10% for practical angles out- or inward on a "normal" Dob), which buries it in the noise (given even a small change in the section of the truss poles will have more effect).
Quote:
the top and bottom Vs carry some of the load.
The glass, though, is only 10% full. I call that empty, and if I want to call that glass full I'll pick a smaller glass, i.e. a tube with a thicker section  .
Quote:
A small angle change makes very little difference, percentage-wise, but it does improve rigidity a little bit.
Exactly.
Quote:
In the big scopes, the angles are often more greatly exaggerated, but then the weights born by the tubes are as well. It doesn't matter if the top is narrower than the bottom or vice-versa. It's the difference that counts.
Yes. And small differences don't really help that much. Only when you stray quite far from the usual compact Dob would the impact be significant (if you manage to angle your trusses 20° or even 10° it'll help a lot, but on my 400mm retorfit I'm building, that either makes the UTA diameter zero or makes the mirror box twice as wide ).
Might as well make a real Serrurier truss with a much wider middle while you're at it (that would certainly help), but your scope will be a lot bigger and more cumbersome.
Quote:
that if there was a slight torsional twist imparted.
Not that easy to design in a system. And certainly not when you're using round tubes, given that before you attach them they'll want to relax all the torsion out of themselves (a round tube is rotationally symmetric). Come to think of it, you could do it if you had square tubes...
I'm wondering what it's doing to the resistance to compression or elongation, which is actually still more important than resistance against sagging per se.
--------------------
400mm f/4.46 self made Dobsonian on Tom Osypowski equatorial platform
Orion Starblast (114mm f/4 reflector, Alt/Az)
|
Houdini
professor emeritus
Reged: 07/13/07
Posts: 525
Loc: Europe
|
|
Yes, we've had this discussion about non-parallel truss poles not so long ago, see this thread. The conclusion was that it makes little difference - for amateur telescopes at most of the order of 10%.
The same can be said about the "slight torsional twist" (whatever is meant by that, the word "torsion" is used very loosely on this forum). A reasonable amount of pretension will prevent slippage of the poles in their clamps (because of friction), and IMHO the potential stiffness improvement throug pretension is probably coming mostly from improved clamping of the truss poles at the mirror box or upper cage.
Robert
--------------------
16" f/4.9 motorized alt-az, 25" f/5 Dobson, 43" f/4 alt-az under construction
Mirror Edge Support Calculator
|
Starman1
Vendor - Scope City
Reged: 06/24/03
Posts: 12230
Loc: Los Angeles
|
|
Guys, you miss the point.
Even if the improvement in stiffness is only 5%, making the poles non-parallel is a "brick in the road" to making the telescope stiff enough to reduce or eliminate collimation changes with altitude.
It's easy to accomplish and costs nothing.
Is it as important as the pole O.D.? No. But that doesn't mean that it has no value--especially when it can be done easily.
I made the attachment points of the top of my poles 2" (50mm) farther out when I rebuilt my UTA. A small difference (100mm larger circle diameter) to be sure, but it helped a bit.
As for torsion in the poles, it, too, was easy to accomplish on my scope since the end pieces on my poles are attached with "star-fangled washers" (i.e.nut inserts) and are flat. They were originally parallel (when the poles were attached there was no torsional twisting of the poles whatsoever. As an experiment, and remembering my racing days, I made the ends non parallel so that a mild torsional twisting was applied to the poles when the knobs were tightened down. Guess what? The side-to-side flexibility of the poles reduced noticeably. It took slightly more effort to flex the poles toward one another.
By grabbing the UTA and shoving the scope from side-to-side rapidly, you can usually see the poles "snake" a bit as the mirror box movement "lags" behind the UTA movement. The worst I've ever seen this was on a (expensive commercial scope--name deleted). After all modifications to my scope this still did not go away completely. Ideally, the scope should be so stiff it moves as it it were one solid piece.
But these are minor effects compared to tightening the spider vanes and securing the primary mirror. Collimation shifts with pointing altitude can largely be eliminated by addressing these two factors.
--------------------
Don Pensack
12.5" Truss Dob, 5" Maksutov, Fujinon Binos
Sustaining Lifetime IDA member
|
skyler
Carpal Tunnel
Reged: 08/16/06
Posts: 1672
Loc: TGPNW
|
|
I am glad this part was addressed since for the majority of users, that 10% or even "marginal" changes may not have much of a consequence. For me, when I build my partial truss scope for mounting on an EQ mount, I will need to have as much rigidity as possible since it will be a potential issue when imaging.
My plan is to over-build so this has been a great topic for me to follow and a real plus for me to hear both sides.
S
|
Scott Watson
sage
Reged: 05/26/06
Posts: 282
|
|
I've had good luck with the following 6-tube arrangement. First, with 6 truss tubes, you get a slightly higher angle per tube than you do with 8. 6 is not only lighter, but has higher truss stiffness. 8 is overconstrained and MUST have some tortional loading. 6 is not and does not. I also don't have true truss mounts at all. All 6 tubes are hard mounted (glued) to the secondary cage and to a 6-sided mounting plate that attaches directly to the primary box with pinned and registered holes. I almost never have to collimate. When I do, it us usually the secondary mirror, not the trusses that is out of whack.
|
sixela
Postmaster
Reged: 12/23/04
Posts: 10874
Loc: Boechout, Belgium
|
|
Quote:
Guys, you miss the point.
Even if the improvement in stiffness is only 5%, making the poles non-parallel is a "brick in the road" to making the telescope stiff enough to reduce or eliminate collimation changes with altitude.
It's easy to accomplish and costs nothing.
The cost depends entirely on the design.
Quote:
I made the attachment points of the top of my poles 2" (50mm) farther out when I rebuilt my UTA. A small difference (100mm larger circle diameter) to be sure,
To some, that's called "cost" . Because of dimensions (my 400mm rebuild is designed to be stored compactly in my car so that I do not have to choose between family and scope), because of the extra weight for the UTA that causes design changes elsewhere, etc.
I don't have 50mm to play with - that's roughly the entire UTA ring width on my 400mm rebuild (and if I moved inside even less than that, light from the UTA pole ends would be able to reach the focal plane).
Even if I had 50mm, the angle would be 2°, over a rectangle so wide it really wouldn't matter.
By all means, *if* you have a design that lets you use non-parallel truss pairs easily, do so.
But that's a big if (given that it is a secondary design consideration), and your original post made it sound as if it were essential for collimation stability.
--------------------
400mm f/4.46 self made Dobsonian on Tom Osypowski equatorial platform
Orion Starblast (114mm f/4 reflector, Alt/Az)
|
Russell Chase
super member
Reged: 05/09/07
Posts: 102
Loc: Herriman, Utah
|
|
Quote:
I built a double-truss 14" Newtonian...supported by 6-pole trusses from a central ring bearing...I was hoping that the sag at both ends would somewhat cancel out, as Ross mentioned with the Serrurier design.
My 18” uses a central ring that supports four 2” parallel struts that pass through the central ring (they are removed during transportation). The lower end also has four 1.25” struts that keeps the mirror box attached when the 2” tubes are removed. I’ve also thought that there might be some sort of counter acting sag on this scope. Given that it is not an engineered truss design, there must be measurable sag; but using an autocollimator, I’m hard pressed to see much if any.
The picture is of my scope when I forgot two of the tubes – I think it is easier to visualize my design with the missing tubes. With only two tubes I did notice sag with a laser, but everything looked good at the eyepiece - once I found something to add weight to the UTA.
|
Dick Jacobson
super member
Reged: 12/22/06
Posts: 132
Loc: Plymouth, Minnesota, USA
|
|
Has anyone tried a balanced secondary assembly? There would be a counterweight on the opposite side of the spider from the secondary. This should eliminate the need for extreme tightness of the spider vanes. The amount of counterweight might be less than the weight of structure needed to support tight vanes. The counterweight could consist of batteries to power an anti-dew heater for the secondary.
--------------------
Equatorial Newtonians, 20" and 14"
"The purpose of life is to be defeated by ever greater things." - Rielke
|
Previous
Index
Next
Threaded
Edit
Reply
Quote
[Re: 